DELIVERY QUALITY OF EXPERIENCE (QoE) IN A COMPUTER NETWORK

1. An apparatus comprising: an interface connectable to a network to monitor a media program streaming across the network to a media client; a video buffer model to estimate buffering events of a video buffer of the media client during the streaming of the media program based on a modeling of the video buffer; and a user model coupled to the video buffer model, the user model to generate a variable signal representative of an estimation of a quality of experience of the media program during playback of the media program at the media client, wherein during an estimated buffering event in the playback the user model is to update the variable signal to reflect monotonically increasing user dissatisfaction and wherein during a recovery region following the estimated buffering event in the playback the user model is to update the variable signal to reflect monotonically increasing user satisfaction.

2. The apparatus of claim 1 , wherein the variable signal is bound by at least one of an upper threshold and a lower threshold.

3. The apparatus of claim 1 , wherein the video buffer model is to change functions used to generate the variable signal based on the estimation of the buffering events.

4. The apparatus of claim 1 , wherein the user model reflects user dissatisfaction due to a duration of stalls of playback of the media program.

5. The apparatus of claim 1 , wherein the user model reflects user dissatisfaction due to a frequency of stalls of playback of the media program.

6. The apparatus of claim 1 , wherein the user model reflects user dissatisfaction due to a location of stalls of playback of the media program.

7. The apparatus of claim 1 , wherein the user model reflects user dissatisfaction due to a both a duration of stalls and a frequency of stalls of playback of the media program.

8. The apparatus of claim 1 , further comprising: a media session model coupled to the video buffer model, the media session model to model a media session of said media program, the modeling of the media session incorporating a plurality of session items corresponding to the streaming of said media program; and wherein the video buffer model estimates buffering events further based on the modeling of the media session.

9. The apparatus of claim 1 , wherein the video buffer model utilizes a plurality of programmable buffer fullness thresholds to estimate buffering events.

10. The apparatus of claim 9 , wherein one or more of the plurality of programmable buffer fullness thresholds is programmed based at least in part on: a web site address; a device type; and a media player type.

11. The apparatus of claim 1 , further comprising: a hypothetical video buffer model to estimate buffering events for a generalized greedy buffer model that is independent of the video buffer model; and an objective session model coupled to the hypothetical video buffer model, the objective session model to generate a variable signal representative of an objective estimation of a quality of experience of the media program during playback of the media program at a hypothetical media client.

12. In an apparatus comprising an interface connected to a network to monitor a media program streaming across the network to a media client, a method comprising: estimating, using a video buffer model, buffering events of a video buffer of the media client during the streaming of the media program based on a modeling of the video buffer; and generating, using a user model, a variable signal representative of an estimation of a quality of experience of the media program during playback of the media program at the media client, wherein during an estimated buffering event in the playback the user model is to update the variable signal to reflect monotonically increasing user dissatisfaction and wherein during a recovery region following the estimated buffering event in the playback the user model is to update the variable signal to reflect monotonically increasing user satisfaction.

13. The method of claim 12 , wherein the variable signal is bound by at least one of an upper threshold and a lower threshold.

14. The method of claim 12 , wherein the user model reflects user dissatisfaction due to at least one of: a duration of stalls of playback of the media program; a frequency of stalls of playback of the media program; and a location of stalls of playback of the media program.

15. The method of claim 12 , further comprising: modeling, using a media session model, a media session of said media program, the modeling of the media session incorporating a plurality of session items corresponding to the streaming of said media program; and wherein estimating buffering events comprises estimating buffering events further based on the modeling of the media session.

16. The method of claim 12 , wherein the video buffer model utilizes a plurality of programmable buffer fullness thresholds for estimating buffering events.

17. The method of claim 16 , further comprising: programming one or more of the plurality of programmable buffer fullness thresholds based at least in part on: a web site address; a device type; and a media player type.

18. The method of claim 12 , further comprising: estimating, using a hypothetical video buffer model, buffering events for a generalized greedy buffer model that is independent of the video buffer model; and generating, using an objective session model, a variable signal representative of an objective estimation of a quality of experience of the media program during playback of the media program at a hypothetical media client.

19. An apparatus comprising: an interface connectable to a network to monitor a media program streaming across the network to a media client; a video buffer model to estimate buffering events of a video buffer of the media client during the streaming of the media program based on a modeling of the video buffer; and a user model coupled to the video buffer model, the user model to generate a variable signal representative of an estimation of a quality of experience of the media program during playback of the media program at the media client, wherein the estimation of the quality of experience reflects user dissatisfaction due to both a location and a duration of each of the estimated buffering events, and wherein in response to the video buffer model indicating an estimated buffering event the user model is to update the variable signal to reflect monotonically increasing dissatisfaction and wherein the user model otherwise is to update the variable signal to reflect monotonically increasing satisfaction.

20. The apparatus of claim 19 , wherein the variable signal is bound by at least one of an upper threshold and a lower threshold.